Significance: This project evaluates access to emergency care for Veterans with particular emphasis on: assessing resources and capabilities of VA EDs to care for Veterans during times of crisis, determining where care is provided (VA vs. non-VA EDs), and understanding how and why Veterans make decisions regarding where to seek emergency care. The objectives of this proposal are directly responsive to HSRD priority areas (Access and Health systems change) and three of the VA Breakthrough priority areas (Improve Access to Health Care, Improve Community Care, and Improve the Veteran Experience). We also address Secretary Shulkin's priority areas (Choice Act, improve infrastructure). Dr. Vashi and her team have strong ties to VA and non-VA operational and policy partners, including the VA National Director of Emergency Medicine, which will be instrumental in translating research findings into meaningful improvements in policy. Background and Innovation: Emergency care is a critical but understudied part of the continuum of health care services offered to Veterans by the VA. While the VA is committed to providing timely and high- quality emergency care, surprisingly little is known about Veteran access to acute care or about the quality of care provided. Variations in VA emergency care resources are not well understood and prevent Veterans from making informed choices. Moreover, eligibility for non-VA ED care is confusing, complex and can translate to delays in treatment and poor outcomes. VA estimates that current VA users get, on average, only about 38% of their ED visits from VA, yet no study to date has examined emergency care Veterans receive in non-VA settings. While use of non-VA EDs may sometimes be appropriate, there are other times when the ED care could have been provided at the VA and use of the non-VA ED may result in duplicative care and delays. This dual use matters to the VA because, like an Accountable Care Organization, the VA is still responsible for the overall health of the Veteran and often the expenses even when Veterans receive care outside of the VA. Aims: Objectives of this proposal are: (1) To create survey items that facilitate a comprehensive inventory of VA ED resources and capabilities; (2) to calculate VA and non-VA ED utilization rates and identify the patient, facility, and community-level predictors of VA ED use and VA reliance; and (3) to characterize Veterans' preferences, resources and contextual factors that influence ED setting choice (i.e. VA, non-VA). Methods: We will build on pilot work and convene an advisory panel to create and add survey items to the next ED survey to assess VA ED resources and capabilities and determine how they vary (e.g. region, rurality, complexity). Using an innovative linkage method, we will use VA, Medicare, and California data to examine ED use across groups. Further, we will assess the patient, facility, and community level predictors of Veterans' choice of delivery system and VA reliance. All analyses will be conducted for three cohorts: a national accountable veteran population (Veterans 65+), residents of California (Veterans 65+), and residents of California (Veterans 18-64). Finally, we will conduct semi-structured interviews with Veterans to better understand why Veterans choose to receive their ED care inside or outside the VA. As underscored by the Secretary, our methods rely on ?what Veterans actually tell us? and lay the groundwork for future comparisons of VA and non-VA ED quality metrics. Expected results: We will determine capabilities of VA EDs nationwide and examine the impact of these factors on VA ED use and reliance. Further by analyzing data and speaking with Veterans, we will better understand where and why Veterans access the acute care system (in both VA and non-VA settings) and will identify barriers in access amendable to future intervention and policy changes. Once we can identify patients receiving care outside the VA and determine why they made that decision, we can then develop interventions to keep Veterans ?in-network? when appropriate.